Fundamental to an efficient episodic memory system is the ability to remember selectively-to be able to focus attention on relevant memories at the exclusion of irrelevant memories. While it is thought that prefrontal cortex (PFC) plays a critical role in selecting among competing memories, there remains considerable ambiguity concerning how competition is represented at a neural level and how PFC interacts with posterior cortical sites to achieve selection. This proposal seeks to advance understanding of how competition impacts processing in the cortical regions that support 'reactivation' of memories and how signals from these regions relate to engagement of specific PFC subregions. The first experiment will assess the representation of memory competition in regions that process perceptual elements of events and are reactivated during retrieval. It is hypothesized that the fidelity of signals in these perceptual regions will be related to PFC engagement. The second experiment will assess the generality of PFC mechanisms that resolve competition by formally testing for overlap between mechanisms that govern competitive retrieval and those that govern perceptual decision-making. Finally, the third experiment will assess the adaptive qualities of retrieval, probing the biases in reactivation that come about as a consequence of prior acts of retrieval. Across each of these experiments, experimental paradigms will be structured such that relevant (target) and irrelevant (distracter) memories are always from visual categories that are processed by distinct perceptual cortical regions (i.e., Faces and Scenes). This approach will allow for direct comparison of reactivation that is related to targets vs. distractors and will therefore be essential for gauging the relationship between PFC engagement and the signals propagating from perceptual cortical regions. These experimental pursuits are intended to form the basis for more applied experimental questions related to the memory failures that are frequently observed in a variety of clinical and non-clinical states. Summary: The present proposal seeks to use functional MRI to better understand how, at a neural level, competition between episodic memories is experienced and resolved. This work underscores the fact that memory often fails not because information is not learned in the first place, but because retrieval operations are overwhelmed by competition. The insights to be gained are of obvious and critical relevance to memory disorders and impairments associated with aging and dementia, particularly of the Alzheimer's type.